Selective detection schemes are of great importance in determining targeted components from complex matrices. This strategy is useful when studying the disposition of drugs, when evaluating new biochemical processes, or when searching for environmental contamination. The chemical reaction interface/mass spectrometer approach (CRIMS) is one of the most powerful methods devised for selectively searching for compounds when the analyst cannot specify their composition more precisely than some unique isotopic or elemental signature. Previous experiments from this laboratory have determined that CRIMS selectively detects 13C, 15N, and deuterium from biological specimens at concentrations at or below 1 ng/mL. These analyses are linear over 3 orders of magnitude. Isotope ratios can be quantified from isotopic dilutions as low as 1:1000. The level of detection of the stable isotopes suggests that CRIMS could eliminate a substantial number of experiments where radioactivity is used as a general detector of metabolic products because stable-isotope CRIMS can do the same thing. Based on this background, the perspectives of this application are a combination of hardware development, chemical innovation, and applications in biomedical research. As its developer, Dr. Abramson proposes a set of experiments which are designed to improve existing capabilities, expand the range of what CRIMS can accomplish, and to test these abilities by selected applications. The technology developments include mechanical measures to make CRIMS yet more sensitive, reliable, and reproducible, yet less expensive. The continued progress towards HPLC interface for CRIMS is a high priority so that this separation technique, which is increasingly important in biomedical analyses, can benefit from the selective detection schemes of which CRIMS is capable. Chemical innovations should expand the selective detection capabilities of CRIMS to include halogens and phosphorus. To evaluate how well CRIMS can assist in biomedical analyses, a selected group of applications are proposed, each using some sort of isotopic or elemental label. These range from the metabolic pattern of 13C-labeled tryptophan to the study of sulfur-linked drug metabolites. In each application, the ability of CRIMS to follow a label in a general, yet sensitive and selective, fashion should produce a result which could not be obtained by other means.